Skip to content

This repo contains my natural language processing projects including the use of neural language models for text generation based on Spinoza's works and a sentiment analysis "in reverse" applied to movie reviews.

Notifications You must be signed in to change notification settings

marcotav/natural-language-processing

Repository files navigation

Natural Language Processing

Natural-language processing (NLP) is an area of computer science and artificial intelligence concerned with the interactions between computers and human (natural) languages, in particular how to program computers to fruitfully process large amounts of natural language data. Challenges in natural-language processing frequently involve speech recognition, natural-language understanding, and natural-language generation.

Notebooks and descriptions Contact Information Notebooks and descriptions Contact Information

Notebooks and descriptions

Notebook Brief Description
neural-language-model-and-spinoza Spinoza's Ethics is used to build a language model for text generation with recurrent neural nets.
sentiment-analysis A "reverse sentiment analysis" using Bernoulli Naive Bayes was performed on movie reviews (already classified) to identify which words appear more frequently on reviews from each class.
topic-identification Tutorial about topic identification (in progress)
alphabet-human-thought/meaning-of-sentences In this notebook, it will be shown that using logic formalisms one can find more generic translation mechanisms (in progress)
alphabet-human-thought/sentence-structure We will show how to develop formal models for patterns in sequence of words using grammars and parsers (in progress)

Contact Information

Feel free to contact me:

About

This repo contains my natural language processing projects including the use of neural language models for text generation based on Spinoza's works and a sentiment analysis "in reverse" applied to movie reviews.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published